๐ง๐ผโโ๏ธ
Our goal for this project was to explore how does an LSTM network work, and try to implement it on a text document. We chose to work with the book โPeter Panโ, by J. M. Barrie. We decided to take the English version, because the English alphabet has fewer characters (only 26 versus 42 in the French one, for instance). Furthermore, the main advantage of working with an official (published) source, is that it eases the cleaning process as there wonโt be spelling mistakes. This means that there is no need to put any correction function in place (as we would have to do if we were working with reviews, or comments from a website, for instance) which lets us focus on LSTMs.
pip install keras
This tutorial assumes a Python 2 or Python 3 development environment with SciPy, NumPy, and Pandas are installed.
The tutorial also assumes scikit-learn and Keras v2.0+ are installed with either the Theano or TensorFlow backend.